The Principal-Agent Model and Regulatory Federalism
| Author | Patricia Metz,Michael J. Sgicchitano,David M. Hedge |
| Published date | 01 December 1991 |
| Date | 01 December 1991 |
| DOI | 10.1177/106591299104400414 |
| Subject Matter | Articles |
THE
PRINCIPAL-AGENT
MODEL
AND
REGULATORY
FEDERALISM
DAVID
M.
HEDGE
and
MICHAEL J.
SGICCHITANO,
University
of
Florida
and
PATRICIA
METZ,
West
Virginia
University
ver
the
past
several
years,
researchers
have
drawn
upon
a
~
principal-agent
model
to
analyze
federal
regulatory
behavior.
Central
to
that
analysis
is
the
assumption
that
federal
bureau-
crats
are
more
responsive
to
elected
officials,
including
presidents,
the
Congress
and
the
courts
than
previously
thought.
Support
for
that
position
has
been
found
in
a
variety
of
policy
settings,
including
the
NLRB
(Moe
1982
and
1985),
OSHA
(Scholz
and
Wei
1986),
and
the
EEOC
(Wood
1990).
Research
by
a
number
of
scholars
suggests
that
a
principal-agent
model
can
also
guide
analysis
of
federal-state
rela-
tions.
Chubb
provides
the
most
cogent
statement
of
that
thesis.
Accord-
ing
to
that
author,
federalism
can
be
viewed
as
a
hierarchal
system
of
control
in
which
federal
officials
are
able
to
obtain
state
compliance
to
national
standards.
Federal
control
is
achieved
in
two
stages.
Initially,
federal
elected
officials
shape
the
decisions
of
federal
bureaucrats,
who,
in
turn,
condition
the
decisions
of
subnational
governments.
Although
the
evidence
(discussed
below)
is
incomplete,
a
handful
of
studies
(Chubb
1985;
Scholz
and
Wei
1986;
Crotty
1988)
suggest
that
the
states
are
responsive
to
federal
signals
even
in
the
wake
of
the
Reagan
administration’s
efforts
at
devolution.
The
viability
of
a
principal-agent
view
of
federalism
is
particularly
important
for
regulatory
federalism.
Many
of
the
social
and
protective
regulatory
programs
enacted
in
the
1960s
and
1970s
established
a
joint
federal-state
regulatory
partnership
that
operates
off
a
common
partial
preemption
framework.
Under
partial
preemption,
federal
author-
ities
outline
a
series
of
national
standards.
The
states
are
then
invited
to
develop
regulatory
programs
consistent
with
those
standards
as
a
condition
for
enforcing
the
program
within
their
state’s
borders.
Until
Received:
June
27,
1990
Revision
Received:
October
30,
1990
Accepted
for
Publication:
November
6,
1990
Note:
The
authors
are
grateful
for
the
constructive
advice
and
criticisms
of
the
editor
and
the
anonymous
reviewers.
1056
state
programs
are
approved,
federal
enforcement
prevails.
Once
state
programs
receive
approval,
federal
authorities
shift
to
a
largely
over-
sight
role
(see
ACIR
1984
and
Wellborn
1988).
Implicit
in
partial
pre-
emption
are
the
notions
of
heirarchy
and
control
central
to
the
principal-
agent
model - state
officials
will
respond
to
the
inducements
of
federal
authorities
in
much
the
same
way
that
federal
bureaucrats
respond
to
members
of
Congress
and
the
White
House.
In
short,
a
test
of
the
principal-agent
model
is
also
a
very
direct
test
of
the
assumptions
that
undergird
regulatory
federalism
in
America.
Despite
the
importance
of
the
issue,
the
evidence
necessary
to
support
(or refute)
a
hierarchal
view
of
federalism
is
incomplete.
To
demonstrate
the
utility
of
a
principal-agent
model
in
explaining
federal-
state
relations,
two
linkages
need
to
be
established,
something
prior
research
has
been
unable
to
do.
First,
analysis
must
demonstrate
that
federal
regulators
are
responsive
to
national
political
principals
(e.g.
Congress,
federal
court
officials,
or
the
President).
Second,
it
must
be
shown
that
federal
regulators
are
able,
in
turn,
to
condition
the
regu-
latory
behaviors
of
state
agencies.
To
date,
however,
no
one
has
tested
both
linkages
within
a
single
context.
Much
of
the
available
research
(e.g.
Moe’s
analysis
of
the
NLRB
or
Wood’s
1988
study
of
the EPA)
is
limited
to
the
national
level
and,
accordingly,
considers
only
the
first
linkage.
Moreover,
only
Scholz
and
Wei
(1986)
and
Hedge
and Jallow
(1990)
look
at
the
first
linkage
in
terms
of
interstate
variations
in
fed-
eral
regulatory
activity
(the
actual
unit
of
analysis
in
Chubb’s
model).
In
a
similar
fashion,
Chubb’s
and
Scholz
and
Wei’s
analyses
examine
the
direct
influence
of
congressional
and
presidential
influences
on
state
behavior
and
ignore
the
mediating
influence
of
federal
bureaucr-
ates - an
influence
that
is
central
to
the
model’s
viability.
Finally,
Crotty’s
analysis
of
EPA’s
relations
with
the
states
only
considers
the
relations
between
federal
and
state
regulatory
agencies
and
ignores
the
first
linkage.
The
issue
of
the
principal-agent
model’s
viability
is
made
more
problematic
by
additional
studies
that
challenge
the
assumptions
of
heirarchy
and
control.
Prior
research
on
federal
grant
programs,
for
instance,
indicates
that
state
and
local
governments
are
frequently
able
to
resist
federal
efforts
to
control
their
behavior
and
frequently
reshape
federal
programs
to
reflect
state
or
local
interests
and
priorities
(see,
for
example,
Ingram
1977
and
Pressman
and
Wildlavsky
1973).
There
is
also
evidence
that
state
officials
are
often
able
to
influence
the
decisions
of
federal
authorities
(see,
e.g.
Haider’s
1974
study
of
To continue reading
Request your trial